Quaternaire Revue de l'Association française pour l'étude du Quaternaire vol. 29/3 | 2018 Volume 29 Numéro 3

Genesis and of the archaeological layers of Pedra Furada -shelter, Genèse et taphonomie des niveaux archéologiques de l’abri de la Pedra Furada, Brésil

Fabio Parenti, Alan Cannell, Evelyne Debard, Martine Faure et Mercedes Okumura

Édition électronique URL : http://journals.openedition.org/quaternaire/10313 ISSN : 1965-0795

Éditeur Association française pour l’étude du quaternaire

Édition imprimée Date de publication : 1 septembre 2018 Pagination : 255-269 ISSN : 1142-2904

Référence électronique Fabio Parenti, Alan Cannell, Evelyne Debard, Martine Faure et Mercedes Okumura, « Genesis and taphonomy of the archaeological layers of Pedra Furada rock-shelter, Brazil », Quaternaire [En ligne], vol. 29/3 | 2018, mis en ligne le 01 septembre 2020, consulté le 10 janvier 2019. URL : http:// journals.openedition.org/quaternaire/10313

© Tous droits réservés Quaternaire, 29, (3), 2018, p. 255-269 255

GENESIS AND TAPHONOMY OF THE ARCHAEOLOGICAL LAYERS OF PEDRA FURADA ROCK-SHELTER, BRAZIL

n

Fabio PARENTI1,2, Alan CANNELL2, Evelyne DEBARD3, Martine FAURE4 & Mercedes OKUMURA5

Pour Claude, un Maître et un ami.

ABSTRACT

New sedimentological and taphonomic analyses of a preserved bulk of Pedra Furada rock-shelter characterize the mixed origin of the illing, both from the sandstone wall and from the uppermost conglomerate layer. Statistical analysis of quantitative and qualitative data on waterfalls and archaeological assemblage, both dated from the Upper Pleistocene and Holocene, demonstrated no causal relationship among Eastern waterfall talus, lithic tools and archaeological structures. Technical similarities have been reported between extant monkeys unintentional laking by-products and the lithic toolkit from the previous archaeological excavations. Comparison between present monkey’s hammerstones and the quartz elements recorded in archaeological structures point to a clear anthropic selection in the making of archaeological structures in the site, regardless of any association with charcoals. Because of the similarities between the present monkey stone-tools and the (supposed) human tool-kit, the inclusion of both human and primate archaeologies in an interdisciplinary research program is needed to clarify the nature of pre-Clovis presence in South America.

Keywords: Sedimentology, Upper Pleistocene, Pedra Furada, statistics, lithic tools, archaeological structures, Platyrrhines

RÉSUMÉ

GENÈSE ET TAPHONOMIE DES NIVEAUX ARCHÉOLOGIQUES DE L’ABRI DE LA PEDRA FURADA, BRÉSIL Les nouvelles recherches sédimentologiques et taphonomiques conduites sur le bloc-témoin de l’abri de la Pedra Furada mettent en évidence la double origine du remplissage, provenant de la paroi gréseuse et des niveaux conglomératiques supérieurs. Les données quantitatives et qualitatives concernant les chutes d’eau et le contenu archéologique du site, tant dans les niveaux pléistocènes que holocènes, démontrent l’absence de relation de causalité entre le talus de la chute d’eau orientale, l’outillage lithique et les structures archéologiques. Des similaritées technologiques ont été observées et comparées entre les produits lithiques non intentionnels des singes actuels et ceux provenants des anciennes fouilles. La comparaison entre la masse des percuteurs utilisés par les singes actuels et celle des galets de quartz utilisés, retrouvés dans les structures archéologiques, indique une évidente sélection d’origine anthropique dans la disposition des structures, avec ou sans restes de combustion. Du fait de la similarité entre l’outillage des singes actuels et celui (supposé) humain dans le passé, une double archéologie anthropique et simienne est le seul moyen d’éclaircir la vraie nature de la présence humaine pré-Clovis en Amérique du Sud.

Mots-clés : Sédimentologie, Pléistocène supérieur, Pedra Furada, statistiques, outillage lithique, structures archéologiques, Platyrrhiniens

1 - INTRODUCTION Meltzer et al., 1994; Borrero, 1995; Guidon et al., 1996; Michab et al., 1998; Parenti, 2001). Since 1974 the entire For more than twenty years, the region of the Serra Serra da Capivara National Park and its surroundings da Capivara National Park, located in Piauí state, have been intensively studied from a multidisciplinary Northeastern Brazil, has been the focus of a harsh approach, recording - among others - several sites debate on pre-Clovis evidence in South America, mainly presenting Pleistocene palaeontological (Guérin & Faure, centered on the evidence of the reference site, the 2008, 2014) and archaeological remains (Lourdeau & Boqueirão da Pedra Furada (BPF) rock-shelter (main Pagli, 2014). In particular, since 2008, research under the references: Guidon & Delibrias, 1986; Butzer, 1988; direction of Eric Boëda has been conducted with the aims

1 Departamento de Antropologia, Universidade Federal do Paraná, PR, BR-80060-150 Curitiba. Email: [email protected] 2 Istituto Italiano di Paleontologia Umana, IT-03012 Anagni. Email: [email protected] 3 Université de Lyon, UCBL, ENSL, CNRS, LGL-TPE, FR-69622 Villeurbanne. Email: [email protected] 4 Université Lumière - Lyon 2, UFR Temps et Territoires; and Université de Lyon, UCBL, ENSL, CNRS, LGL-TPE, FR-69622 Villeurbanne. Fundação Museu do Homem Americano, São Raimundo Nonato (Piauí, Brasil). Email: [email protected] 5 Laboratório de Estudos Evolutivos Humanos, Departamento de Genética e Biologia Evolutiva, Instituto de Biociências, Universidade de São Paulo, Rua do Matão, 277, Cidade Universitária, Butantã, SP, BR-05508-090. Email: [email protected] Manuscrit reçu le 31/10/2017, accepté le 26/02/2018 256 of replicating and reining the data obtained at the BPF site there has recently been the discovery of platyrrhine and discovering new evidence of Pleistocene occupations monkeys as battering and laking agents in the Brazilian in the region (Lahaye et al., 2013; Boëda et al., 2014, lowlands (Profitt et al., 2016). 2016). Although the results obtained at BPF, excavated In the background of the criticisms advanced about between 1978 and 1988, have been exhaustively published the Pleistocene sites in Serra da Capivara region, we since 2001 (Parenti, 2001; Santos et al., 2003; Valladas et are well aware that the core issue is the taphonomic and al., 2003; Chaves et al., 2006), many recent publications contextual analysis of each speciic set of evidence. For a do not even consider the abovementioned detailed data, long time the focus of research has been the chronology preferring to focus their critiques on a generic and and classiication of archaeological remains, without conjectural set of taphonomic and technological remarks or paying due attention to the formation processes of even, naively, ignoring the existence of the site itself when deposits and evaluation of alternative explanations. In considering the issue of a pre-Paleoindian occupation in this sense, a irst step was made in the book on BPF, South America (Bueno et al, 2013; Vialou et al., 2017). when considering gravity as a possible agent for cobbles Meanwhile, several research programs have focused (Parenti, 2001, ch. 5). However, at the time on complementary aspects of the main debated issue, i.e. of this study (1988-1992) a detailed sedimentological the hypothesis of a human colonization of the Americas analysis was still missing and therefore a well-founded during the Pleistocene; among these are: 1) a huge statistical analysis could not be carried out. array of genetic studies that aim at tracing the origin of In this article we present four new sets of data concerning population stock(s) and the timing of their spread(s), the formation processes at BPF: 1) a sedimentological generally admitting a irst peopling event not older description of the illing, partially based on the 1987- than 20 ka BP (Zegura et al., 2004) 2) a model of two 1988 excavations but mainly on unpublished data from settlement waves based on cranial morphology (Hubbe a ield campaign conducted in 1996 on the preserved et al., 2014); 3) new archaeologically consistent and bulk in the Eastern sector of the site; 2) a statistical well-dated pre-Clovis sites (McAvoy & McAvoy, 1997; comparison between the natural fracture of quartz cobbles Stanford et al., 2014 ; Dillehay et al., 2015); 4) the role from the base of the western waterfall and the artifacts of Beringia refugia during the Last Glacial Maximum as recovered in the excavations, in order to investigate any cradle of American population waves (Hoffecker et al., gravitational action in the making of the assemblage 2016); 5) a re-birth of the controversial hypothesis of and its relation to the artefacts, manuports or supposed an North-Atlantic Ice-Age coastal entry route in North geofacts; 3) a comparison between the granulometry of America, i.e. the “” (Stanford & the illing and the composition of some representative Bradley, 2012; Eren et al., 2013). To add to this debate features of Pleistocene units; 4) a synopsis of main

Fig. 1: Position of the site (BPF) in relation to the sandstone cuesta and the nearby rock-shelter of Sítio do Meio (SDM) ; image from GoogleEarth. The whitish edge of the cliff is the uppermost outcropping conglomerate layer. Note the thalweg just South of BPF, comprised between elevations 383 and 393. Fig. 1 : Localisation du site (BPF) par rapport à la cuesta gréseuse et à l’abri-sous-roche du Sítio do Meio (SDM) ; image de GoogleEarth. Le bord blanchâtre de la falaise est le niveau conglomératique le plus élevé. À noter que le talweg immédiatement au Sud de BPF est compris entre les élévations 383 et 393. 257 technical parameters observed on lithic tools compared both with waterfall breakage and with available data of unintentional laking made by monkeys.

2 - SEDIMENTOLOGY OF THE FILLING

The Boqueirão da Pedra Furada rock-shelter (8°51’S and 42°33’200’’W) lies at the bottom of the Serra Grande Silurian formation with a 70 m high cuesta (igs. 1 & 2). The lower part of the cliff is carved from medium to coarse grained siliceous, low cemented, sandstone, presenting a slightly extended cross-bedding stratiication (maximum thickness 80 cm). Some iner, centimetric, silty layers are interbedded at the bottom. The whole formation dips north-northwest between 8 and 15º. These sandstones are overburdened by a widely cross-bedded conglomerate (ig. 2). Pebbles and cobbles are heterometric and normally graded. The largest are found in the lowermost meters and reach a length of 30-40 cm. Their composition is heterogeneous: quartz, quartzite, sandstone and occasionally basic rocks. Progressively there are medium to coarse grey sandstones, with conglomeratic lenses. Several systems determine the orientation of main drainage axis and cuesta dissection. The shelter is formed by differential erosion between sandstone and silty layers, engendering a 19 m overhang at the reference section (ig. 2, section II-II’). At the bottom of the drip line a talus points to the last retreat of the cliff, responsible for the overhang formation and closing the shelter to South. Four waterfalls drain the uppermost conglomerate, the main one affecting the site Fig. 2: Sections of the sandstone wall (top) and situation of being waterfall C (ig. 2 & 3). The shelter substratum has a waterfalls and excavations in 1990 (bottom). 10° east-west dip, according to the general sandstone dip. Fig. 2 : Coupes de la paroi gréseuse (en haut), et localisation des chutes d’eau et de la fouille de 1990 (en bas).

Fig. 3: Top: position of waterfalls, excavations, sections and structures discussed in this paper; bottom: I) Panoramic view at the end of excavations (July 1988) with sections discussed in this paper, II) Northern half of section 3 during excavations (November 1987), III) section of waterfall C; note the block collapse beneath cobbles illing. Fig. 3 : En haut : situation des chutes d’eau, des fouilles, des coupes et des structures dont il est question dans cet article ; en bas : I) Vue panoramique à la in de la fouille (juillet 1988) avec les coupes présentées dans cet art icle, II) Moitié Nord de la coupe 3 pendant la fouille (novembre 1987), III) Coupe au niveau de la chute d’eau C ; à noter l’effondrement d’un bloc au-dessous des galets du remplissage. 258

2.1 - SEDIMENTOLOGICAL ANALYSIS Joël Pellerin in 1988 and published in Parenti (2001). Three sedimentary groups had been deined at that time, The chronostratigraphic reference during the excavations taking into account their structure and alteration. and the one utilized for the regrouping of cultural evidence Sediment samples of the 1996 campaign were taken is the central section (Parenti, 2001, pl. 8). close to the wall for section 4 and underneath the drip line In 1996, eight years after the last excavation, the best for section 20. Moreover, in order to know sedimentary preserved portion of the illing was the Eastern bulk variation westward, three samples were taken on section (ig. 3). It lies on a sandstone wall and is the reference for 3’ (ig. 6), 1 m east and below section 3 (ig. 3). the present stratigraphic description (tab. 1), especially Elements larger than 3 cm were never a majority in the for sections 4 (ig. 4) and 20 (ig. 5); the latter showing analyzed samples (ig. 7), these comprise between 5 and the deposit’s evolution outside the drip line. 25 % of the samples from section 20, yet are completely A preliminary sedimentary study was carried out by absent or scarcely represented in section 4. The only

Layers Stratigraphic description Thickness Colour Lateral and vertical variations Remarks coarse sand containing small quartz pebbles 1 - 5 dark yellowish present only in Northern part of 1 (1-2 cm) et some sandstone plates cm brown (10YR 4/2) the bulk highly reworked moderate 2 coarse sand with two thin lenses of max : yellowish brown close to the wall, gravel and small highly reworked by ants granules and gravels 14 cm (10YR 5/4) pebbles are more abundant

lens rich in small pebbles (1-2 cm) and max: dark yellowish clearly thicker in its Northern part pebbles are imbricated, mostly latly 3 round gravel 15 cm orange (10YR 6/6) disposed against the limits the area closer to the wall is highly bioturbated by ants. Rich 4a sand surronding isolated small pebbles 30 - 40 grayish orange coarser. Lesser amount of matrix in large and dispersed charcoal cm (10YR 7/4) and some imbricated pebbles in fragments, sometimes clustered as sectors attained by runoff in a ireplace small sandstone plates, associated with 4b scattered pebbles (1-5 cm), in a rich sandy 10 - 35 grayish orange largest plates, horizontally resting, form some large quartz cobbles matrix cm (10YR 7/4) two lines on the edges of the layer (10x15 and 6x10 cm) in its lower part, some sandstone sandstone plates and quartz pebbles in a 40 - 50 grayish orange blocks and large siliceous cobbles one large block diving into the lower 4c rich sandy matrix containing many gravels cm (10YR 7/4) (7 cm wide), most of the time layer and small pebbles horizontally disposed angular granules, isolated or aligned in small lenses in a very rich grey-orange 40 - 50 grayish orange 4d matrix. Gravels and abraded small pebbles cm (10YR 7/4) (1-3 cm), with no preferential orientation

sand rich in fragmentary charcoals visible all 5 - 20 light brown general dip to NW. On section 20 5 along its height. Small sandstone plates and (ig. 5), coarser sand with sparse lower limit is irregular small quartz pebbles (1-4 cm) were observed cm (5YR 6/4) cobbles (max 4 cm) more small sandstone plates close rich sandy matrix, with granules, gravels 20 - 25 grayish orange to the wall and at the opposite (in 6 and scattered small pebbles. cm (10YR 7/4) Southern portion) more siliceous gravels and small pebbles sandy matrix rich in micro-charcoals and grayish orange 7 charcoals homogeneously distributed max : (10YR 7/4) to dark along its thickness. Sandstone plates 15 cm yellowish orange (5 cm long) are plentiful and often altered (10YR 6/6) similar to layer 6 with abondant sandy from the drip line outward there is a matrix surronding granules and small 30 - 35 grayish orange rise in pebble length to some 7 cm. 8 pebbles without preferential orientation in cm (10YR 7/4) They are distributed in two beds on relation to layer’s boundaries. the top and the bottom of the layer. Southward, it vanishes and is visible charcoal layer, more widespread than only by the presence of isolated 5 - 20 grayish orange charcoals among many small 9 the others, but with the same stochastic cm (10YR 7/4) distribution of charcoals pebbles. In section 20 (ig. 5) it is reduced to a small charcoal lens. similar to preceding layers : iner granulometry 10 in Northern part, coarser in the South, with 20 - 25 grayish orange many pebbles 1-5 cm long cm (10YR 7/4) very coarse, with abundant quartz pebbles associated with plates and small blocks 30 - 35 light brown to the South, increase in number of sandstone elements are altered and 11 of sandstone. Large charcoal fragments cm (5YR 5/6) pebbles but not in their size friable scattered among pebbles. sandy matrix surrounding small elements 12 (granules, gravels, small pebbles 1-2 cm). 12 - 15 grayish orange Southward it covers the blocks at Micro-charcoals. cm (10YR 7/4) the bottom of section 20.

Large blocks in its upper part the layer envelops a talus formed in a 13 very coarse layer with large pebbles 40 cm dark yellowish pointing to the last roof collapse in (6-7 cm) associated with sandstone plates orange (10YR 6/6) the Eastern sector of the shelter. previous collapse with a 40° dip to NW ine layer : sand and most elements less dark yellowish ills the gaps between the substrate 14 than1 cm orange (10YR 6/6) and the collapsed blocks Tab. 1: Stratigraphy of the Eastern bulk Tab. 1 : Stratigraphie du témoin Est. 259

Fig. 4: Section 4, western side of preserved bulk. Fig. 4 : Coupe 4, côté ouest du bloc-témoin.

Fig. 5: Section 20, southern side of preserved bulk. Fig. 7: Comparison of all globlal granulometries. Note its position slighty outside the drip line (ig. 3). Fig. 7 : Diagrammes comparatifs des granulométries globales. Fig. 5 : Coupe 20, côté sud du bloc-témoin. À noter sa position légèrement en dehors de la ligne d’égouttement (ig. 3). contribution of silt than in the ones more distant from the wall (section 20, ig. 5). The ones close to the wall exceptions are layers 4a and 4b with 10-15 % of coarse are relatively homogeneous except in layer 7, which is elements. These are also present in layer 1 of section 3’. iner. Sediments in section 20 are more heterogeneous. The bulk of the sediment is composed by particles The sample from layer 4 in section 3’ is clearly different smaller than 2 mm (igs. 8 & 9). These are mostly because it contains 30 % of silt. A sample from the illing coarse sands with a 15-20 % of silt and very scarce clay below waterfall B, draining the upper conglomerate, (2-3 %). Close to the wall, the samples present a greater contains only sands (silt contribution is less than 5 %).

Fig. 6: Section 3’, eastern side of central preserved bulk. This section is 1 m east and below section 3. Fig. 6 : Coupe 3’, côté est de la partie centrale du bloc-témoin (ig. 3). Cette coupe se trouve 1 m à l’est et en contre-bas de la coupe 3. 260

Fig. 8: Selected granulometric curves from sections 4, 3’, Waterfall B, section 20. Fig. 8 : Sélection de courbes granulométriques cumulatives des coupes 4, 3’, chute d’eau B, coupe 20.

In general, the sedimentological analysis shows the difference between samples lying just below the drip line (section 20) - coarser and heterogeneous - and the ones well inside the sheltered area (sections 3’ and 4), which are iner and more homogeneous. The latter, however, present some variability due to the sandstone lenses from which, partly, they are derived. The siltiest sample from section 3’ shows a strong contribution of ine sediments due to runoff at the center of the sheltered area. Some layers (layer 3 for example) are clearly thicker in Northern part of the bulk and could be the result of a higher runoff along the sandstone wall. Field observations and granulometric analysis conirm the twofold origin of the deposit, as previously proposed by Guidon and Parenti (1989). The bulk of the ine elements is derived from the disintegration of the sandstone wall, as the weak cementation of the majority of its layers might contribute to this process. Although continuous, the main causal agents might have changed over time, including chemical alteration during the moister phases, haloclasty and/or wind erosion during the drier periods. In some layers rich in charcoals (layer 7 for example), sandstone plates are often altered: apart from pedological process, their alteration may be due to heating. Fig. 9: Structure 67, PF3 phase, and ireplace 19 (19 300 ± 200 BP). The presence of sandstone slabs and blocks points to the Fig. 9 : Structure 67, phase PF3, et foyer 19 (19 300 ± 200 BP). peeling phases of the wall. Their nature varies according 261 to the weathered layers of the wall: coarse sandstone, the three Talus cones (A, B, C, ig. 2) (Parenti, 2001, rich in iron oxides or without limonite, ine yellowish p. 144). In summary, during the 1987-1988 excavations, sandstone. This is particularly clear for the sediments of any artifacts in the sheltered area that complied with the the northern sector, where sandstone fragments may be criteria of at least one clear (i.e. with a concave surface related to rip scars still visible on the wall. It is the same and evident point of impact) lake-scar for laked pieces for the eastern block (mostly on its Eastern part) as well (cores or choppers) and, for detached pieces, the presence as for the central bulk (section 3’). of an evident butt and bulbar surface were collected. Siliceous coarser elements (measuring more than Here we statistically compare the stones from Talus cone 1 cm) are derived from conglomeratic layers; these A (considered as presenting a natural, non-anthropogenic were deposited during events of wetter periods or origin) and the BPF excavated material (considered simply thunderstorms which stimulated the activity as having an anthropogenic origin). The BPF sample of the waterfalls. For the sectors exposed in this includes materials both from Holocene (Agreste and study, presenting the maximum concentration of Serra Talhada) and from Pleistocene (Pedra Furada) archaeological structures (eastern sector and Trench 6), units (ig. 4). Our hypothesis is that there are important the bulk of cobbles and pebbles is derived from differences between cobbles from Talus cone A and BPF waterfall C and, partially, from minor drains eroding artefacts, the latter presenting characteristics that are the wall upon the shelter. different from stones broken by gravity (Talus cone A). The sample from Talus cone A consisted of complete 2.2 - DEPOSITION DYNAMICS and broken cobbles, natural lakes, and fragments (deined as pieces of rock that do not present common Just after the last recession of the wall (older than 60 ka lake features, like a bulb, butt or a sharp edge). Elements greater than 32mm were suitable for analysis. The BP; Parenti, 2001, pl. 8 and Santos et al., 2003), which originated the block collapse closing the shelter to the archaeological pieces consisted in cores and lakes, either retouched or not. south, the irst sediments, derived from the wall and from The following variables were analyzed for broken the upper conglomerate illed up the existing hollows according to the dip both of talus and substratum. These pebbles (Talus cone A) and laked pebbles or cores sediments have been accompanied by block collapses (excavation): number of bulb scars larger than 1 cm : 1, 2 inducing the wall regularization. The presence of large or > 3; maximum lenght, in mm: 32-64, 64-128, 128-256 sandstone slabs in some layers points to the spheroidal (Wentworth, Krumbein modiied, granulometric classes); of harder layers, favorable to an erosion pseudo-retouch: present or absent. The variables analyzed in “onionskin”. Sedimentation continued on with a for lakes were : laking angle: > 90° or ≤ 90°; cortex: granulometric gradient from east to west and from south present or absent; non cortical butt: present or absent; to north for the elements originated from conglomerates pseudo-retouch: present or absent. and from north to south for the elements delivered from We assume that human laking would result in cores the wall; the original NW dip decreased along time. presenting a greater amount of lake scars and pseudo- In this monotonous sedimentary dynamic, there are retouch, non-cortical lakes, non-cortical butt and laking layers of coarser cobbles, pointing to moister episodes angle greater than 90°. Z-tests for the equality of two which enhance the potential of creeks and waterfalls, as proportions were applied using 5 % signiicance level well as indicating the erosion of coarser lenses inside the in order to test the presence of signiicant differences conglomerate. between proportions of lake and core attributes between Talus cone A materials and archaeological artifacts. Some cobbles observed in the sheltered area could not have been naturally deposited. The regular structure of Table 2 presents the number of bulb scars larger than layers and the lack of channel beds rule out any torrential 1 cm observed in broken or laked pebbles for Talus cone event that could have contributed to the deposit of A and the archaeological layers. Materials from Talus cobbles inside the shelter. No geodynamic agent, water cone A do not present more than four lake-scars, whereas or gravity, could have transported large elements, such archaeological materials present the full range of such as the big cobble (22 x 14 cm) found in section 3’ feature (data not shown in detail due to z test limitations). Pleistocene and Holocene materials presented no among iner elements and with a reduced dip (ig. 6). An anthropic (or other animal?) intervention must therefore be assumed to explain the presence of such large cobbles Talus Archaeological Layers Variable cone A within the shelter. Pleistocene Holocene 1 65 (81.2) 8 (4.6) 12 (5.3) Number of bulb-scars >1 15 (18.8) 165 (95.4) 216 (94.7) 3 - COMPARISON BETWEEN THE STONES DEPOSITED BY WATERFALLS Total 80 173 228 AND THE ARTEFACTS. Tab. 2: Number of bulb-scars observed in broken or laked pebbles from Talus cone A and archaeological layers (% in parentheses). Tab. 2 : Nombre d’esquillements observés sur les galets cassés ou A irst comparison between the two sets was established débités du talus du cône A et des couches archéologiques (% entre on the basis of the observation of 2,000 cobbles from parenthèses). 262 signiicant differences between the proportion of the Table 5 shows that there is a very low number of number of bulb scars (one and more than one, z-score natural lakes with a striking angle greater than 90° when -0.2911, p-value 0.77182), therefore, it was possible to compared to the absolute numbers in the archaeological sum up these numbers when comparing to Talus cone material from the Pleistocene (there are no data from A. There were signiicant differences in the proportion Holocene materials). The z-test points to a signiicant of the number of bulb scars between Talus cone A and difference between the proportional natural and the archaeological layers (z-score 16.3284, p-value 0). In archaeological lakes regarding the laking angle (z-score fact it is possible to observe that 81.2 % of pebbles from -6.7087, p-value 0). Talus cone A present only one bulb scar, while the same attribute can be observed in around 5 % of Pleistocene Variable Talus cone A Pleistocene and Holocene cores. Flaking angle > present 3 (10.7) 85 (78.7) 90° Granulometric information regarding materials from absent 25 (89.3) 23 (21.3) Talus cone A and the archaeological site is presented in Total 28 108 table 3. Due to limitations of the z-test, data from 64-128 and 128-256 mm was summed up, creating a new interval Tab. 5: Number of lakes presenting striking angles greater than 90° from Talus cone A and archaeological layers (% in parentheses). from 64 to 256 mm. Interestingly, there were signiicant Tab. 5 : Nombre d’éclats du talus du cône A et des couches differences between the proportions of the samples from archéologiques présentant un angle de frappe supérieur à 90° (% entre Talus cone A (with and without lake scars, z-score parenthèses). 2.7071, p-value 0.00672) and between the Pleistocene Table 6 presents the number of lakes according to and Holocene materials (z-score -3.5687, p-value the presence of cortex. There is a complete absence of 0.00036). The comparison between proportions of Talus natural lakes (from Talus cone A) presenting cortex cone A (without lake scars) and Pleistocene revealed and a great proportion of lakes presenting cortex in the signiicant differences (z-score 5.218, p-value 0). The archaeological layers. However, the proportion between other comparisons (Talus cone A without lake scars and Pleistocene and Holocene materials is signiicantly Holocene, Talus cone with lake scars and Pleistocene, different (z-score 3.3619, p-value 0.00078), as well as and Talus cone with lake scars and Holocene) did not the proportion between Talus cone A and Pleistocene present signiicant differences between proportions. lakes (z-score -11.9462, p-value 0) and Talus cone A and Holocene lakes (z-score -15.4117, p-value 0). 32-64 mm 64-128 mm 128-256 mm Total

Talus No bulb-scars 171 (45.1) 200 (52.8) 8 (2.1) 379 Archaeological layers cone Variable Talus A cone A With bulb-scars 24 (28.9) 55 (66.3) 4 (4.8) 83 Pleistocene Holocene Pleistocene 40 (22.2) 125 (69.4) 15 (8.4) 180 present 0 112 (97.4) 1598 (86.6) Holocene 94 (38.5) 138 (56.6) 12 (4.9) 244 Cortex absent 42 (100) 3 (2.6) 247 (13.4) Tab. 3: Number of elements according to granulometric range in broken or laked pebbles from Talus cone A and archaeological Total 42 115 1845 layers (% in parentheses). Tab. 3 : Nombre d’éléments en fonction de la granulométrie des galets Tab. 6: Number of lakes presenting cortex from Talus cone A and cassés ou débités du cône A et des couches archéologiques (% entre archaeological layers (% in parentheses). parenthèses). Tab. 6 : Nombre d’éclats corticaux du talus du cône A et des couches archéologiques (% entre parenthèses).

Table 4 shows that the observed proportion of pseudo Given that the proportions observed for the lakes retouch in broken pebbles from Talus cone A and presenting non cortical butt do not show signiicant laked pebbles from the site are very low. Due to the differences between Pleistocene and Holocene contexts lack of signiicant differences between the proportions (z-score 0.151, p-value 0.88076), these values were observe for Pleistocene and Holocene materials (z-score summed up to be compared to the values from Talus cone -0.2424, p-value 0.81034), these values we summed A (tab. 7). There is a complete absence of lakes from Talus up and compared to the values observed in Talus cone cone A presenting non cortical butt, while the opposite A, showing signiicant differences between the natural pattern can be observed in the archaeological materials. materials and the archaeological ones (z-score -2.0384, The z-test shows that there are signiicant differences p-value 0.04136). between such proportions (z-score -5.222, p-value 0).

Archaeological layers Talus cone A Pleistocene Holocene Variable Talus cone A Pleistocene Holocene present 2 (0.4) 3 (1.7) 5 (2.0) Pseudo- Non present 0 50 (40.3) 767 (39.6) retouch absent 461 (99.6) 175 (98.3) 244 (98.0) cortical butt absent 42 (100) 74 (59.7) 1168 (60.4) Total 463 178 249 Total 42 124 1935 Tab. 4: Number of broken or laked pebbles presenting pseudo- retouch from Talus cone A and archaeological layers ( % in parentheses). Tab. 7: Number of lakes presenting non cortical butt from Talus Tab. 4 : Nombre de galets cassés ou débités du talus du cône A et des cone A and archaeological layers (% in parentheses). couches archéologiques (% entre parenthèses) présentant des pseudo- Tab. 7 : Nombre d’éclats non corticaux du talus du cône A et des retouches. couches archéologiques (% entre parenthèses). 263

Table 8 presents the number of lakes according to the presence of pseudo retouch. Both the Talus cone A and the archaeological layers present a very low frequency of lakes where retouch can be observed. However, there are signiicant differences between the proportions observed in the Pleistocene and Holocene materials (z-score 12.4613, p-value 0), as well as between the Talus cone A and Pleistocene artifacts (z-score -2.4353, p-value 0.01468). No signiicant differences were observed between Talus cone A and Holocene materials (z-score 1.9512, p-value 0.05118).

Talus Archaeological layers Variable cone A Pleistocene Holocene present 5 (11.9) 53 (30.5) 103 (5.1) Pseudo- retouch absent 37 (88.1) 121 (69.5) 1910 (94.9) Total 42 174 2013 Fig. 10: Selected structures from Pleistocene layers. Tab. 8: Number of lakes presenting pseudo-retouch from Talus 69 (PF2)/ small structure close to the sandstone wall, composed by cone A and archaeological layers (% in parentheses). large (apparently unheated) quartz cobbles, none of them having traces Tab. 8 : Nombre d’éclats du talus du cône A et des couches archéologiques of pitting; 23 (PF3)/ ireplace composed by sandstone blocks and quartz présentant des pseudo-retouches (% entre parenthèses). pebbles, with charcoal concentration in the central concavity; 30 (PF2)/ ireplace boarded by rounded sandstone blocks and four quartz cobbles; When comparing the pieces from Talus cone A and note the ine gravelly matrix of the sediment. from the archaeological layers, it is possible to state that Fig. 10 : Sélection de structures des couches pléistocènes. 69 (PF2)/ petite structure proche de la paroi gréseuse, faite de grands galets en there are signiicant differences between the frequencies quartz (apparemment non chauffés), aucun ne présente de traces de observed for the analyzed variables in natural pieces pitting ; 23 (PF3)/ foyer fait de blocs de grès et galets de quartz, avec une concentration de charbon au centre de la concavité ; 30 (PF2)/ foyer and in the ones we assumed anthropogenic (from the bordé de blocs de grès arrondis et de quatre galets de quartz ; à noter la archaeological layers). Such differences in the proportion ine matrice de graviers du sédiment. of features between the natural and archaeological materials can be observed in most analysed variables: from Pedra Furada 3 phase, we unearthed structure 67, number of bulb scars (tab. 2) and the presence of pseudo a very large stone pile, composed of 515 cobbles and retouch in pebbles or cores (tab. 4), as well the presence 102 sandstone blocks, elongated in north-south direction of a striking angle greater than 90° in lakes (tab. 5), and bordered on its Western side by a darkened circular and presence of non-cortical butt in lakes (tab. 7). For area ( 19, ig. 9) presenting charcoals dated at some other variables, the differences between the natural 19,300 ± 200 BP (GIF 8125). Such structure was found and archaeological materials were not so clear, but still on the inner portion of the shelter, with a slight N-S dip, could be observed in some comparisons (tabs. 3, 6, as observed in ig. 3-II. & 8). The important differences between natural and Besides the presence of these abovementioned archaeological materials revealed by the statistical structures that very likely resulted from anthropogenic analysis supports our hypothesis that endogenous raw deposition, a dimensional analysis of a cross-section materials from both Pleistocene and Holocene layers from N-S was also performed on structure 67. A total from BPF are of a different origin (anthropogenic) than of 617 cobbles, presenting a maximum dimension of natural materials from Talus cone A. 240 mm were analyzed (ig. 11). Not only there is a

4 - THE FILLING AND THE ARCHAEOLOGICAL STRUCTURES

For illustrating the relationship between archaeological features and the sedimentary matrix of the illing, we chose some well-deined stone structures from Pleistocene layers, as described in Parenti (2001). All the selected features (igs. 9 & 10) are not only well inside the drip line, but also in upslope position, frontally opposed to the talus slope, as shown in igure 3. The lithological composition of these structures, i.e. the proportion between quartz cobbles and sandstone slabs is variable, but all their elements are concentrated and clearly larger than the sedimentary matrix, excluding Fig. 11: Distribution of cobbles from structure 67 by maximum length (2 cm intervals). the possibility of deposition due to hydraulic transport Fig. 11 : Répartition des galets de la structure 67 en fonction de leur from eastern waterfall. Among the relevant structures longeur maximum (intervalles de 2 cm). 264 weak correlation (r = 0.49) between the size of the stones behaviour regarding tool use (Visalberghi et al., 2007; and the distance from the actual waterfall, but their Mannu & Ottoni, 2009; Falótico, 2011; Falótico & distribution is random, with no apparent differentiation Ottoni, 2013; Visalberghi & D. Fragaszy, 2013; Falótico by slope (ig. 12); moreover, the whole structure is & Ottoni, 2016). In parallel a new body of research has dipping opposite to the talus slope (ig. 3II). That means been developed in order to study material remains of we can reject the hypothesis that these stones were non-human primate’s instrumental and social activities, carried and sorted by gravity, related to the distance from leading to the creation of a true non-human primate the waterfall. The distribution of these stones, according , irstly developed on the present African to maximum length (grouped by 2 cm intervals) is shown Pan territory (Mercader et al., 2002; Carvalho et al., in ig. 11. There is a marked preference for stones with 2008; Carvalho & McGrew, 2012) and after also applied a maximum length of between 8 and 10 cm, followed in northeastern Brazil and notably the Serra da Capivara by those of 6 to 8 and 10 to 12 cm. The mean maximum region (Haslam et al., 2016). length of these stones was 92 mm, which corresponds to However, the dificulty is the scarcity of primate fossil a mass of about 400 g for quartz cobbles, based on data remains which often prevent any reliable attribution between average mass and lengths for cobbles (Falótico of authorship of a speciic artefact assemblage. In the & Ottoni, 2016). As indicated below, this distribution Serra da Capivara region, despite the copious Upper around a mean of 400 g is typical of anthropic selection Pleistocene fossil assemblage consisting in more than (Cannell, 2002, 2018). 7000 faunal remains of about 60 species proceeding from 12 paleontological sites (Guérin & Faure, 2008), monkey fossil remains are absent except for one tooth of Alouatta from Toca do Barrigudo site (Guérin & Faure, 2014). Alouatta sp. has been also reported in the Toca dos Ossos cave, Ourolandia, Bahía state (Auler et al., 2006). Also in Bahia state a new species of Alouatta, A. mauroi, has been described in the late Upper Pleistocene site of Gruta dos Brejões (Morro do Chapéu) (Tejedor et al., 2008). Two fossilized specimens of Atelinae were discovered in 1992 in the Pleistocene megafauna of Toca da Boa Vista site (Campo Formoso, Bahia), about 300 km SE of Serra da Capivara Park. One of the skeletons has been attributed to Protopithecus brasiliensis Lund, 1838, deined in the fossiliferous deposits of Lagoa Santa, Minas Gerais state, Brazil (Hartwig & Cartelle, 1996). The other was described as a new genus and new species Caipora Fig. 12: Structure 67: maximum cobbles dimension (cm) by distance bambuiorum Cartelle & Hartwig, 1996, an Atelinae from waterfall C (m). Distances are irregularly spaced because of the possibly reaching a body mass of 20 kg. Recently Halenar different quantity of cobbles for each meter considered. Fig. 12 : Structure 67 : dimension maximale des galets (en cm) en (2012) reviewed Protopithecus brasiliensis, and Halenar fonction de la distance de la chute d’eau C (en mètres). Les distances & Rosenberger (2013) reassessed the almost complete sont irrégulièrement espacées du fait de la différente quantité de galets pour chaque mètre pris en compte. skeleton from the Toca da Boa Vista deining it as a new genus and new species Cartelles coimbrailhoi Halenar & Rosenberger, 2013, presenting a body mass of about 5 - MONKEYS, ARTEFACTS AND 25-28 kg. STRUCTURES AT BPF Beside these discoveries, paleontological data are meager (Macphee & Horovitz, 2002). Without going In the last decades cognitive and instrumental into the details of the Cebidae phylogenetic relationship capabilities of non-human primates have been largely and dispersal, the ancestor of the Cebinae presently studied, leading to a review of the former more living in northeastern Brazil is unknown. Ruiz Ramoni anthropocentric view asserting that tool use is a unique et al. (2017) recently published the discovery of a Plio- human feature (Moura & Lee, 2004; Haslam et al., 2009; Pleistocene molar tooth of Cebidae from Venezuela, being Visalberghi et al., 2015). A large body of research has “the irst deinite evidence of capuchine monkeys in the documented material cognition and tool use in Old South America fossil record”. We do not yet know when World Hominid, but less attention has been payed to Cebinae colonized northeastern Brazil, so it is impossible the instrumental behavior of platyrrhine monkeys of the to conirm their presence in Upper Pleistocene. Haslam et New World until the end of 20th century. Stone-tool using al., (2016) dated stone anvils probably used by Cebidae behavior in captive Cebus / Sapajus is systematically some 600-700 years ago in the Serra da Capivara Park; documented from the irst half of 20th century onwards this Pre-Columbian age was immediately quoted by (Visalberghi, 1990; Westergaard & Suomi, 1994). The use Fiedel (2017) but it is irrelevant to say that the Cebinae of tools by capuchin monkeys in the caatinga and cerrado lived in this region in the Upper Pleistocene. However, environment of Southern Piauí has been intensely studied fossil Atelidae are well attested. At present, Cebinae since at least 2007, showing a very rich and diversiied are the only known platyrrhine monkeys using stone 265 hammers. Do they have the monopoly of this behavior? PF1 PF2 PF3 ST1 ST2 AG Oit Oit We know that in Africa several catarrhine primates (early surf TOT Hominins, Homo, Apes) had, and have, this behavior; Choppers+cores 66 72 40 115 104 25 10 21 so we should wonder if in South America the large- Av. weight 419 504 464 442 284 494 130 204 sized extinct Atelinae (Protopithecus, Caipora and/or Av. mx. Length 97 99 95 84 84 86 65 72 Cartelles) also showed this behavior; demonstrating this % of cortical coverage >50% 73 81 90 60 48 56 82 would be a great advance in knowledge of the anthropoid Complete lakes 26 42 15 889 876 145 19 31 platyrrhine evolution. Av. weight 66 67 94 35 35 55 15 20 Currently capuchin monkeys frequently feed at the Tab. 9: Comparison between Pedra Furada site (endogenous raw bottom of sandstone wall, where cobbles for smashing material) and Capuchin monkey’s materials. nuts of various species are very common; they also Total number of pieces is reported on the right (Profitt et al., 2016). repeatedly smash quartz cobbles in outcropping Tab. 9 : Comparaison entre les artefacts du site de la Pedra Furada (matériaux bruts endogènes) et ceux des singes capucins. Le nombre conglomerate layers inside the sandstone cuesta, total de pièces est indiqué à droite (Profitt et al., 2016). apparently in order to produce a desirable powder smell (Mannu & Ottoni, 2009; Falótico & Ottoni, 2016; Profitt assemblages at BPF also shows striking similarities, et al., 2016); bang stones as auditory behaviour (Moura, strongly suggesting that there was a technological 2007), as well as use stones as digging tools for the continuum in endogenous raw material selection and extraction of underground food (Falótico et al., 2017). In tool preparation (ig. 13). these activities they use quartz cobbles as hammerstones, sometimes unintentionally producing cores and lakes, some of them with human-like intentional laking features as: 1) concave negative scars on cobbles, 2) angle between platform and ventral face > 90, resembling that produced in human laking, 3) lake-scars on dorsal face of lakes (Profitt et al., 2016). As very correctly stated by Fiedel (2017) in a recent comment on the Profitt et al. (2016) publication, the stone remains produced by these primates can mimic the simplest elements of the lithic assemblage recovered in Fig. 13: Mass distribution of choppers, cores and lakes in BPF layers. Fig. 13 : Répartition des galets aménagés, nucléus et éclats dans les couches the Serra da Capivara region, in both the Pleistocene and de BPF en fonction de leur masse. Holocene layers. Because equiinality of unintentional stone laking by neo-tropical primates is an important Beside stone-tool using in multiple tasks by monkeys, novelty, the authors of this article consider that the entire we should also consider the possibility that these animals lithic production on endogenous raw material from could have produced rock piles or cairns similar to those this region should be re-considered in the light of this previously deined as anthropic structures in BPF. Stone important paradigm shift. throwing from proceptive females was reported by In order to distinguish between the monkey’s and the Falótico and Ottoni (2013) as a “provocative” behavior (supposed) anthropic lithic collection of Pedra Furada, during estrus, but no stone accumulation or storage an initial comparison between the two assemblages places related to capuchine monkeys have been currently is presented. The bulk of the Pedra Furada toolkit observed in Serra da Capivara. is composed of very simple trimmed cores, quartz However, when considering the possibility that fragments with macroretouches and few clear lakes. In monkeys could have contributed to the site formation, this comparison only data collection from the surface some points concerning the lithic assemblage of was used (Lasca OIT Surface; Proitt et al., 2016), as the (Pleistocene and Holocene) units at BPF should be kept possibility that their excavated artifacts had an anthropic in mind : origin is not - in our opinion - completely ruled out in the a) Unambigous hammerstones have not been recovered Serra da Capivara region. Thus the sample from OIT is in any of the Pleistocene BPF layers; several putative composed of 60 stones, divided as follows (Proitt et al., specimens have one or more, not spatially constrained, 2016, Extended Data tab. 1) (tab. 9): incipient hertzian cones on cortical surface (Parenti, The most evident difference between the two sets 2001, plates 57, 61, 67, 85), but none exhibited repeated, is their mass: archaeological artifacts are two or three crushed and concentrated micropitting on a restricted times heavier than the monkey artifacts for cores surface. These are the features commonly attributed and laked pieces in general and three to four times to (and used to deine) hammerstones and are clearly for lakes. Holocene archaeological lakes are on the associated to percussive activities of monkeys (Haslam average lighter because of the intense knapping of cores et al., 2014; Profitt et al., 2016, Extended Data ig. 2). (ig. 13). It is also important to note that the average mass Thus, the isolated impact points on some core or chopper values of archaeological choppers and cores lie mostly at BPF were attributed to the stochastic falls of cobbles around the 300-500 g mark, similar to Pleistocene and at the surface of waterfalls, i.e. the lithic raw material modern artefact values (Cannell 2012, 2018). The source, before its selection, transport and trimming. mass distribution of both the Pleistocene and Holocene Moreover, the typical pitted depressions resulting from 266 repeated palm nut cracking percussions (Visalberghi et disposal on the waterfall talus. Hence, their use as al., 2007) on the top of the sandstone blocks excavated hammerstones would have implied a minimum transport inside the shelter were never observed. of about 10 m. Thus, a more probable work site for the b) Although dorsal scars on lakes allow us to monkey smashing activity would have been at the large discriminate between waterfall and intentional trimming, sandstone blocks on the talus of waterfall C (ig. 2 & 3), such criteria does not apply to the discrimination between but in this case any by-products would have probably human and non-human laking. For this reason the rolled downstream, possibly “polluting” the outer section secondary use of lakes (retouching or scraping/ cutting of the site, named Vale da Pedra Furada (Parenti, 2015). activities), not observed so far in wild monkeys, can be d) We agree that the multifunctional nature of the a very important factor in ascertaining the nature of a observed monkey tool-kit is a true mimesis of a simple given assemblage. The many lakes and quartz fragments core and lake technology and, because of this, the with marginal retouch recovered from well inside the anthropic origin of a single (simple) artefact cannot be rock shelter and dated from the Upper Pleistocene and precisely ascertained, also taking in account the low Holocene (Parenti, 2001, pl. 70-76, 87-93, 96-97) ought sedimentation rate recorded at BPF (av. 1 cm/century), to be carefully reconsidered and analyzed; as an example which allows for the multiple reutilization of the same we present here three quartz lakes with marginal retouch tool (potentially by monkeys or humans). Therefore from PF1 and PF2 layers (ig. 14). a given assemblage would only be recognizable - on statistical basis - after a careful analysis of chaînes- opératoires and use-wear studies. However we still consider that the most of the more elaborate tools from BPF assemblage can be regarded as the result of intentional laking by humans. Five trimmed cobbles are shown here as examples: two choppers from the oldest layer Pedra Furada 1 (ig. 15) and three cores from the richest Pleistocene layer, Pedra Furada 2 (ig. 16). All have technical features that could hardly be attributed to gravitational fracture or the occasional percussion by monkeys.

Fig. 14 : Retouched and utilized lakes. 14492/ partially cortical quartz lake; retouched on both faces, PF1; 18602/ cortical lake, quartz, marginal retouch on dorsal face, invasive retouch on ventral face, PF1; 17960/ partially cortical quartzite lake; at least three lake scars are present on its dorsal face which, on its right side, is interested by a lat diaclase surface which allows a very good grip with the right hand, explains the notch visible on side view, PF2. Fig. 14 : Éclats retouchés et utilisés. 14492/ éclat partiellement cortical en quartz ; retouché sur les deux faces, PF1 ; 18602/ éclat cortical, quartz, retouche marginale sur la face supérieure, retouche envahissante sur la face inférieure, PF1 ; 17960/ éclat partiellement cortical en quartzite; au moins trois enlèvements sur la face supérieure qui, du côté droit, présente Fig. 15: Bifacial quartzite chopper, with a sinusoidal distal edge, un plan de clivage qui permet une très bonne prise de la main droite, ce formed by alternate trimming, 790 g, PF1 (17883); bifacial qui explique la coche visible sur la vue de côté, PF2. quartzite chopper, with continous cutting edge, deined by many lake-scars, 995 g, PF1 (18637). c) It is worth noting that at BPF itself the conglomerate Fig. 15 : Galet aménagé bifacial en quartzite, avec un bord distal sinusoïdal, obtenu par retouches alternes, 790 g, PF1 (17883) ; galet presenting the cobbles used by monkeys is not exposed aménagé bifacial en quartzite, avec un bord tranchant continu, formé and (loose) cobbles are only available for the monkey’s par de nombreuses traces d’enlèvements, 995 g, PF1 (18637). 267

and the artefacts recovered and considered as having an anthropic origin, both from the Holocene and Pleistocene layers of the site. This comparison allow us to ascertain that part – although not the totality - of the archaeological assemblage could be the result of monkey’s activity, but also that some important technical traits of the artefacts along with the evidence for several stone structures still present a very strong case for a long-lasting pre-Clovis human presence in this region. We are perfectly aware that Pedra Furada - as all its analogs – needs to be studied in its archaeological and natural context and not only with a site-centered focus; it is known that BPF is not an isolated case, as our previous works and ongoing research are demonstrating. From the time when it was dismissed by leading archaeologists as a possible collection of geofacts - without any careful examination of the evidence – Pleistocene layers of BPF have been ignored even by scholars (academically) debating on pre-Clovis peopling of lowland South America. In addition, ethologists are now trying to produce evidence for a pre-Columbian presence of monkey activity in order to develop the exciting new ield of non-human primates archaeology in the Neotropic. These studies should consider existing data Fig. 16: Centripetal quartzite core with minimum four lake-scars, the on archaeological and stratigraphic context as a useful last has been retouched, maybe indicating a last use as scraping tool, 765 g, PF2 (17090); small quartzite core, useless as hammer because tool for further discussion on this interesting research of its weight (155 g) and form, PF2 (4510); orthogonal core on quartz topic. If there is indeed any evidence of a long-lasting cobble, 360 g, PF2 (4430-1); orthogonal core on quartzite cobble; lower view: double synchronous split fracture, due to percussion, overlapping co-existence of humans and monkeys in the 265 g, PF2 (15891). Note the recurrent technique of the last two cores. same region, it is time to build a well-founded research Fig. 16 : Nucléus centripète en quartzite avec un minimum de quatre strategy including both non-human primate and human enlèvements, le dernier a été retouché, cela indique peut-être la archaeology in order to achieve a full understanding dernière utilisation comme grattoir, 765 g, PF2 (17090) ; petit nucléus en quatzite, utilisé comme percuteur du fait de son poids (155 g) et de of past cultural and ecological adaptations in lowland sa forme, PF2 (4510) ; nucléus orthogonal sur galet de quartz, 360 g, South-America. PF2 (4430-1) ; nucléus orthogonal sur galet de quartzite ; vue du bas : double fracture synchrone, obtenue par percussion, 265 g, PF2 (15891). À noter la technique récurrente des deux derniers nucléus. AKNOWLEDGEMENTS 6 - CONCLUSIONS We are very grateful to Walter Neves and Tiago Falótico In this paper we revised the formation of Pedra for suggestions and improvements of the manuscript. Furada deposit, highlighting the main physical agent Tomos Profitt and Tiago Falótico kindly provided us seen as the principal alternative to the anthropic origin with unpublished quantitative data on Serra da Capivara of endogenous artefact assemblages, i.e. the waterfall. monkey’s stone-tools. Two anonimous reviewers have A detailed sedimentological study carried out on the provided useful suggestions. preserved sections eight years after the last excavation is FP and MO hold a CNPq Research Productivity also presented. This study has analytically demonstrated Scholarship (304394/2015-7 and 302163/2017-4). what had already been observed during the excavations, Sedimentological campaign and palaeontological namely: the mixed origin of the illing, both from the research were funded by Fundação Museu do Homem roof and from the waterfall, and the decrease of the Americano, São Raimundo Nonato, Piauí, Brazil. presence of cobbles inside the shelter. On these basis we compared the granulometry of the Western talus A with that of the lithic assemblage and of some important stone structures from Pleistocene units, demonstrating that both did not result from the waterfalls but - on the contrary - were selected, transported, trimmed (if stone tools) or disposed (if bordering possible ) well inside the sheltered area. Given the well-studied use of stone-tool by wild capuchin monkeys in the Pedra Furada area, we show also an initial comparison between the unintentional cores and lakes currently produced by these primates 268

REFERENCES GUÉRIN C. & FAURE M., 2014 - Paleontologia da região do Parque nacional Serra da Capivara. In A.-M. Pessis, G. Martin & N. Guidon (eds.), Os biomas e as sociedades humanas na pré-história da AULER A.S., PILÓ L.B., SMART P.L., WANG X., HOFFMANN região do parque nacional Serra da Capivara, Brasil, vol. II-A. A&A D., RICHARDS D.A., EDWARDS R.L., NEVES W.A. & Comunicacão,, São Paulo, 136-182. CHENG H., 2006 - U-series dating and taphonomy of Quaternary GUIDON N. & DELIBRIAS G., 1986 - Carbon-14 dates point to man vertebrates from Brazilian caves. Palaeogeography, Palaeoclimato- in the Americas 32,000 years ago. Nature, 6072, 769-771. logy, Palaeoecology, 240, 508-522. GUIDON N. & PARENTI F., 1989 - Toca do Boqueirão da Pedra BOËDA E., CLEMENTE-CONTE I., FONTUGNE M., LAHAYE Furada: Escavações 1987. C., PINO M., DALTRINI FELICE G., GUIDON N., HOELTZ S., In IV Reunião Cientíica da Sociedade de Arqueologia Brasileira. Centro Cultural de Santos - 20 a 26 de LOURDEAU A., PAGLI M., PESSIS A.M., VIANA S., COSTA setembro de 1987. Dédalo, 1. Publicações Avulsas, São Paulo, 57-67. A. & DOUVILLE E., 2014 - A new late Pleistocene archaeological sequence in South America: the Vale da Pedra Furada (Piauí, Brazil). GUIDON N., PESSIS A.M., PARENTI F., FONTUGNE M. & Antiquity, 88, 927-955. GUÉRIN C., 1996 - Pedra Furada in Brazil, and its “presumed” evidence: limitations and potential of the available data. Antiquity, BOËDA E. ROCCA R., COSTA A. DA, FONTUGNE M., HATTÉ 70, 416-421. C., CLEMENTE-CONTE I., SANTOS J. C., LUCAS L., FELICE G., LOURDEAU A., VILLAGRAN X., GLUCHY M., RAMOS HALENAR L.B., 2012 - Paleobiology of Protopithecus brasiliensis, a M. P., VIANA S., LAHAYE C., GUIDON N., GRIGGO C., PINO Plus-Size Pleistocene Platyrrhine from Brazil. PhD thesis, The City M., PESSIS A. M., BORGES C. & GATO B., 2016 - New Data on University of New York, New York, 240 p. a Pleistocene Archaeological Sequence in South America: Toca do HALENAR L.B. & ROSENBERGER A.L., 2013 - A Closer Look Sítio do Meio, Piauí, Brazil. PaleoAmerica, 2 (4), 286-302. at the Protopithecus Fossil Assemblages: New Genus And Species BORRERO L.A., 1995 - Human and natural agency: some comments From Bahia, Brazil. Journal of Human Evolution, 65, 374-390. on Pedra Furada. Antiquity, 69, 602-603. HARTWIG W.C. & CARTELLE C., 1996 - A complete skeleton BUENO L., SCHMIDT DIAS A. & STEELE J., 2013 - The Late of the giant South American primate Protopithecus. Nature, 381, Pleistocene/Early Holocene archaeological record in Brazil: a geo- 307-311. referenced database. Quaternary International, 301, 74-93. HASLAM M., CARDOSO R.M., VISALBERGHI E. & BUTZER K.W., 1988 - A Marginality Model to Explain Major Spatial FRAGASZY D., 2014 - Stone Anvil Damage by Wild Bearded and Temporal Gaps in the Old and New World Pleistocene Settle- Capuchins (Sapajus libidinosus) during Pounding Tool Use: A Field ment Records. Geoarchaeology, 3 (3), 193-203. Experiment. PLoS ONE, 9 (11), e111273. CANNELL A., 2002 - Throwing Behaviour and the Mass Distribution HASLAM M., HERNANDEZ-AGUILAR A., LING V., CARV- of Geological Hand Samples, Hand Grenades and Olduvian Manu- ALHO S., TORRE I. DE LA, DESTEFANO A., DU A., HARDY ports. Journal of Archaeological Science, 29, 335-339. B., HARRIS J., MARCHANT L., MATSUZAWA T., MCGREW W., MERCADER J., MORA R., PETRAGLIA M., ROCHE H., CANNELL A., 2018 - Mass distribution analysis of spheroid Manu- VISALBERGHI E. & WARREN R., 2009 - Primate archaeology. ports, spheroid artifacts, and the lithics of play learning. Lithic Tech- Nature, 460, 339-344. nology, 43 (3), 141-150. HASLAM M., LUNCZ L. V., STAFF R. A., BRADSHAW F., CARTELLE C. & HARTWIG W.C., 1996 - A new extinct primate OTTONI E. B. & FALÓTICO T., 2016 - Pre-Columbian monkey among the Pleistocene megafauna of Bahia. Brazil. Proceedings of tools. Current Biology, 26, R515 R522. the National Academy of Sciences, 93, 6405-6409. HOFFECKER J.F., ELIAS S. A., O ROURKE D.H., SCOTT G.R. CARVALHO S., CUNHA E., SOUSA C. & MATSUZAWA T., & BIGELOW N.H. - 2016 - Beringia and the Global Dispersal of 2008 - Chaînes opératoires and resource-exploitation strategies in Modern Humans, Evolutionary Anthropology, 25, 64-78. chimanzee (Pan troglodytes) nut craking. Journal of Human Evolu- tion, 55, 148-163. HUBBE M., OKUMURA M., BERNARDO D.V. & NEVES W.A. 2014 - cranial morphological diversity of Early, Middle, and Late CARVALHO S. & MCGREW W.C., 2012 - The origins of the Holocene Brazilian groups: implications for Human dispersion in : Why chimpanzees (Pan troglodytes) still are good Brazil. American Journal of Physical Anthropology, 155 (4), 546-558. models for technological evolution in Africa. In M. Domín- guez-Rodrigo (ed.), Stone tools and fossil bones: debates in the LAHAYE C., HERNANDEZ M., BOËDA E., FELICE G. D., archaeology of human origins. Cambridge University Press, GUIDON N., HOELTZ S., LOURDEAU A., PAGLI M., PESSIS Cambridge, 201-221. A.M., RASSE M. & VIANA S. 2013 - Human occupation in South America by 20,000 BC: the Toca da Tira Peia site, Piauí, Brazil. CHAVES S.A. DE MIRANDA & REINHARD K.J., 2006 - Crit- Journal of Archaeological Science, 40 (6), 2840 2847. ical analysis of coprolite evidence of medicinal plant use, Piauí, Brazil. Palaeogeography, Palaeoclimatology, Palaeoecology, 237, LOURDEAU A. & PAGLI M., 2014 - Indústrias líticas pré-históricas 110-118. na região da Serra da Capivara. In A.-M. Pessis, G. Martin & N. Guidon (eds.), Os Biomas e as sociedades humanas na pré-história DILLEHAY T., OCAMPO C., SAAVEDRA J., SAWAKUCHI da região do Parque Nacional Serra da Capivara, Brasil, vol. II-A. A. O., VEGA R. M., PINO M., COLLINS M.B., SCOTT L., A&A Comunicacão, São Paulo, 551-635. CUMMINGS L.S., ARREGUI I., VILLAGRAN X. S., HART- MANN G.A., MELLA M., GONZ LEZ A. & DIX G., 2015 - New MACPHEE R.D.E. & HOROVITZ I., 2002 - 13- Extinct Quater- Archaeological Evidence for an Early Human Presence at Monte nary platyrrhines of the Greater Antilles and Brazil. In W.C. Hartwig Verde, Chile. PLoS ONE, 10 (12), e0145471. (ed.), The Primate fossil Record. Cambridge Studies in Biological & Evolutionary Anthropology, 33. Cambridge University Press, EREN M.I., PATTEN R.J., O BRIEN M.J. & MELTZER D.J., Cambridge, 189-201. 2013 - Refuting the technological cornerstone of the Ice-Age Atlantic crossing hypothesis. Journal of Archaeological Science, MANNU M. & OTTONI E.B., 2009 - The Enhanced Tool-Kit of Two 40, 2934-2941. Groups of Wild Bearded Capuchin Monkeys in the Caatinga: Tool FALÓTICO T., 2011 - Uso de ferramentas por macacos-prego Making, Associative Use, and Secondary Tools. American Journal (Sapajus libidinosus) do Parque Nacional Serra da Capivara – PI, of Primatology, 71, 242-251. Tese de Doutorado do Instituto de Psicologia, Universidade de São MCAVOY J.M. & MCAVOY L.D., 1997 - Archaeological Investiga- Paulo, São Paulo, 170 p. tions of Site 44SX202, Cactus Hill, Sussex County, Virginia. Research FALÓTICO T. & OTTONI E., 2013 - Stone Throwing as a Sexual Report Series, 8. Virginia Department of Historic Resources, Rich- Display in Wild Female Bearded Capuchin Monkeys, Sapajus libidi- mond, 487 p. nosus. PLoS ONE, 8 (11), e79535. MELTZER D.J., ADOVASIO J.M. & DILLEHAY T.D., 1994 - On FALÓTICO T. & OTTONI E., 2016 - The manifold use of pounding a Pleistocene human occupation at Pedra Furada, Brazil. Antiquity, stone tools by wild capuchin monkeys of Serra da Capivara National 68, 695-714. Park, Brazil. Behaviour, 153, 421-442. MERCADER J., PANGER M. & BOESCH C., 2002 - Excavation FALÓTICO T., SIQUEIRA J.O. & OTTONI E.B., 2017 - Digging of a chimpanzee stone tool site in the African rainforest. Science, up food: excavation stone tool use by wild capuchin monkeys. Scien- 296, 1452-1455. tiic Reports, 7, 6278. MICHAB M., VALLADAS H., FROGET L. & MERCIER N, 1998 FIEDEL S.J., 2017 – Did Monkeys Make the Pre-Clovis Pebble Tools - Distinguishin gburnt from partly bleached unburnt quartz pebble of of Northeastern Brazil? PaleoAmerica, 3 (1), 6-12. Pedra Furada, Brazil. Ancient TL, 16 (1), 5-9. GUÉRIN C. & FAURE M., 2008 - La biodiversité‚ mammalienne MOURA A.C. DE A. & LEE P.C., 2004 - Capuchin stone tool use in au Pléistocène supérieur - Holocène ancien dans la Région du Parc Caatinga dry forest. Science, 306 (5703), 1909. National Serra da Capivara (SE du Piauí, Brésil). FUMDHAMentos, MOURA A. C. DE A., 2007 - Stone Banging by Wild Capuchin Monkeys: Revista da Fundação Museu do Homem Americano, 7, 80-93. An Unusual Auditory Display. Folia Primatologica, 78, 36-45. 269

PARENTI F., 2001 - Le gisement quaternaire de la Pedra Furada VALLADAS H., MERCIER N., MICHAB M., JORON J.-L. (Piauí, Brésil) : stratigraphie, chronologie, évolution culturelle. REYSS J.-L. & GUIDON N., 2003 - TL age-estimates of burnt Editions Recherches sur les Civilisations, Paris. 323 p. quartz pebbles from the Toca do Boqueirão da Pedra Furada (Piauí, PARENTI F., 2015 - Old and new on the same site: putting Vale da Northeastern Brazil). Quaternary Sciences Reviews, 22, 1257-1263. Pedra Furada into a wider context. A comment to Lahaye et al. 2015. VIALOU D., BENABDELHADI M., FEATHERS J., FONTUGNE Quaternary Geochronology, 30, 48-53. M. & VILHENA VIALOU A., 2017 - Peopling South America’s PROFFITT T., LUNCZ L.V., FALÓTICO T., OTTONI E. B., centre; the Late Pleistocene site of Santa Elina. Antiquity, 91 (358), TORRE I. DE LA & HASLAM M., 2016 - Wild monkeys lake 865-884. stone tools. Nature, 539, 85-88. VISALBERGHI E., 1990 - Tool use in Cebus. Folia Primatologica, RUIZ RAMONI D., RINCON A.D., SOLORZANO A. & MOYÀ- 54, 146-154. SOLÀ S., 2017 - The irst fossil Platyrrhini (Primates: Anthropoidea) VISALBERGHI E. & FRAGASZY D., 2013 - The Etho-Cebus from Venezuela: A capuchin monkey from the Plio-Pleistocene of El Project: Stone-tool use by wild capuchin monkeys. In C.M. Sanz, J. Breal de Orocual. Journal of Human Evolution, 105, 127-131. Call & C. Boesch (eds.), Tool use in animals. Cognition and Ecology. SANTOS G.M., BIRD M.I., PARENTI F., FIFIELD L.K., GUIDON Cambridge University Press, Cambridge, 203-218. N. & HAUSLADEN P.A., 2003 - A revised chronology of the lowest VISALBERGHI E., FRAGASZY D., OTTONI E., IZAR P., occupation layer of Pedra Furada Rock Shelter, Piauí, Brazil: the OLIVEIRA M.G. DE & ANDRADE F.R.D., 2007 - Character- Pleistocene peopling of the Americas. Quaternary Science Reviews, istics of hammer stones and anvils used by wild bearded capuchin 22, 2303-2310. monkeys (Cebus libidinosus) to crack open palm nuts. American STANFORD D. & BRADLEY B.A., 2012 - Across Atlantic ice: The Journal of Physical Anthropology, 132, 426-444. origin of America’s . University of California Press, VISALBERGHI E., SIRIANNI G., FRAGASZY D. & BOESCH Berkeley, 336 p. C., 2015 - Percussive tool use by Taï Western chimpanzees and STANFORD D., LOWERY D., JODRY M., BRADLEY B.A., KAY Fazenda Boa Vista bearded capuchin monkeys: a comparison. Philo- M., STAFFORD T.W. & SPEAKMAN R.J., 2014 - New Evidence sophical Transactions - Royal Society. Biological Sciences, 370, for a Possible Occupation of the Eastern North American 1682, 20140351. Continental Shelf at the Last Glacial Maximum. In A.M. Evans, J.C. WESTERGAARD G.C. & SUOMI S.J., 1994 - A simple stone-tool Flatman & N.C. Flemming. (eds.), Prehistoric Archaeology on the technology in monkeys. Journal of Human Evolution, 27, 399-408. Continental Shelf : a global review. Springer Science, New York, ZEGURA S.L., KARAFET T.M., ZHIVOTOVSKY L.A. & 73-93. HAMMER M.F., 2004 - High-Resolution SNPs and Microsatel- TEJEDOR M.F., ROSENBERGER A.L. & CARTELLE C., 2008 lite Haplotypes Point to a Single, Recent Entry of Native American - Nueva especie de Alouatta (Primates, Atelinae) del Pleistoceno Y Chromosomes into the America. Biology and Evolution, 21 (1), Tardío de Bahía, Brasil. Ameghiniana, 45 (1), 247-251. 164-175.